UARS data and results from a three‐dimensional transport model have been used to compare and contrast the extent of the early stages of chemical processing by polar stratospheric clouds (PSCs) during the 1991–1993 northern and southern hemisphere winters of the Upper Atmosphere Research Satellite (UARS) mission. The modeled location and timing of regions of polar processing agree quite well with the UARS microwave limb sounder (MLS), cryogenic limb array etalon spectrometer (CLAES) and Halogen Occultation Experiment (HALOE) observations in that enhanced ClO concentrations are generally found in sunlit regions of polar‐processed air and decreases of ClONO2 and HNO3 accompany low temperatures and high values of the aerosol extinctions inside the vortex. For these winters it is found that there is a period of about 2 weeks during the northern hemisphere early winter where sporadic polar processing takes place. After this a persistent processing region is seen inside the polar vortex, and the vortex fills with polar‐processed air in a bit more than 2 weeks from the first persistent PSC occurrence. In the southern hemisphere, polar processing also fills the vortex in a bit over 2 weeks. Estimates of the spectral aerosol measure of the aerosol spectrum from CLAES observations illustrate that PSC particles are seen where cold temperatures occur inside the polar vortex and heterogeneous conversion of chlorine species on PSCs are expected.

The ''4-day wave'' is an eastward moving quasi-nondispersive feature with period near 4 days occurring near the winter polar stratopause. This paper presents evidence of the 4-day feature in Microwave Limb Sounder (MLS) temperature, geopotential height, and ozone data from the late southern winters of 1992 and 1993. Spacetime spectral analyses reveal a double-peaked temperature structure consisting of one peak near the stratopause and another in the lower mesosphere, with an out-of-phase relationship between the two peaks. This double-peaked structure is reminiscent of recent three-dimensional barotropic/baroclinic instability model predictions and is observed here for the first time. The height variation of the 4-day ozone signal is shown to compare well with a linear advective-photochemical tracer model. Negative regions of quasigeostrophic potential vorticity (PV) gradient and positive Eliassen-Palm flux divergence are shown to occur, consistent with instability dynamics playing a role in wave forcing. Spectral analyses of PV derived from MLS geopotential height fields reveal a 4-day signal peaking near the polar stratopause. The three-dimensional structure of the 4-day wave resembles the potential vorticity ''charge'' concept, wherein a PV anomaly in the atmosphere (analogous to an electrical charge in a dielectric material) induces a geopotential field, a vertically oriented temperature dipole, and circulation about the vertical axis.

Water vapor in the upper troposphere has a significant impact on the climate system. Difficulties in making accurate global measurements have led to uncertainty in understanding water vapor's coupling to the hydrologic cycle in the lower troposphere and its role in radiative energy balance. The Microwave Limb Sounder (MLS) on the Upper Atmosphere Research Satellite is able to retrieve water vapor concentration in the upper troposphere with good sensitivity and nearly global coverage. An analysis of these preliminary retrievals based on 3 years of observations shows the water vapor distribution to be similar to that measured by other techniques and to model results. The primary MLS water vapor measurements were made in the stratosphere, where this species acts as a conserved tracer under certain conditions. As is the case for the upper troposphere, most of the stratospheric discussion focuses on the time evolution of the zonal mean and zonally varying water vapor. Stratospheric results span a 19‐month period and tropospheric results a 36‐month period, both beginning in October of 1991. Comparisons with stratospheric model calculations show general agreement, with some differences in the amplitude and phase of long‐term variations. At certain times and places, the evolution of water vapor distributions in the lower stratosphere suggests the presence of meridional transport.

Space‐time analyses, which are sensitive to details of retrieval and gridding processes not seen in zonal and time means, are used to investigate the integrity of version 8 gridded retrieved temperatures from the improved stratospheric and mesospheric sounder (ISAMS) on the upper atmosphere research satellite (UARS). This note presents results of such analyses applied to ISAMS tropical data. Comparisons are made with microwave limb sounder (MLS), also on UARS, temperatures. Prominent zonal wave number 1 features are observed with characteristics similar to those expected for Kelvin waves. Time versus longitude plots reveal quasi‐regular eastward phase progression from November 1991 to mid‐January 1992. The perturbations extend throughout the upper stratosphere and lower mesosphere (altitudes of 32–64 km), exhibiting peak‐to‐peak amplitudes of up to 2°–3° K and periods from ∼ 2 weeks in midstratosphere to ∼ 1 week at higher altitudes. Faster Kelvin waves with periods of 3–5 days are also found in the lower mesosphere. Height versus time plots reveal downward phase and upward group velocities, consistent with forcing from below. Vertical wavelengths are ∼ 20 km for the slower mode and about twice this scale for the faster 3 to 5‐day mode. The features are trapped within ±10°–15° of the equator. Kelvin wave signatures in ISAMS and MLS temperatures are compared at 10 and 1 hPa. Good agreement is found, illustrating the internal consistency and ability of both ISAMS and MLS temperature grids to capture relatively small amplitude features with space‐time scales of fast, zonally asymmetric equatorial modes.

We present an example of observations within a single air mass of ClO, ClONO2, and HCl from the instruments on the UARS spacecraft. A three‐dimensional chemistry‐transport calculation for HCl is used to show that this air mass has been transported through regions cold enough for polar stratospheric cloud formation and chlorine activation through heterogeneous chemical reactions. These data, together with modeling of HCl with and without heterogeneous loss as a qualitative measure of polar processing, indicate directly the transformation of the reservoirs HCl and ClONO2 to reactive chlorine species.

UARS Microwave Limb Sounder data and data from the UK meteorological Office indicate that the equatorial zonal mean temperature exhibits a cold anomaly symmetric about the equator, typically lasting 10-20 days and with changes of about 12 K.

The effect of early winter stratospheric minor warmings on the development of the polar vortex is discussed, for winters since the launch of the Upper Atmosphere Research Satellite (UARS).

Satellite observations of ozone and chlorine monoxide in the Arctic lower stratosphere during winter 1992-1993 are compared with observations during other winters, observations of long-lived tracers and the evolution of the polar vortex. Chlorine in the lower stratospheric vortex during February 1993 was mostly in chemically reactive forms.

Satellite observations of ozone and chlorine monoxide concentrations during winter 1992-1993 show that in February 1993 chlorine in the lower stratosphere was mostly in chemically reactive forms. Decreases in stratospheric ozone concentration during February and early March 1993 are consistent with chemical destruction by this reactive chlorine. Comparison with changes in the distribution of long-lived chemical and dynamical tracers shows that the observed decrease cannot have been caused solely by dynamical processes.

Two years of stratospheric measurements of ozone from the Upper Atmosphere Research Satellite Microwave Limb Sounder are examined in order to characterize large horizontal scale wave variations. The use of Fourier analysis allows the detection of variations from daily through seasonal and interannual timescales. High-latitude winter variations at 10 hPa often have very large amplitudes, but smaller midlatitude variations are more ubiquitous. Some variations have the characteristics of locally generated instabilities. Correlations of wave features with changes in the zonal-mean ozone at 10 hPa suggest the presence of significant horizontal motions during strong wintertime polar warming events. Such correlations are not evident at other levels. Spectral analysis of the large-scale variations show most waves to be slowly propagating. In contrast to some past observations, equatorial regions are shown to lack large amplitude wave events.

Global ozone observations from the Microwave Limb Sounder (MLS) aboard the Upper Atmosphere Research Satellite (UARS) are presented, in both vertically resolved and column abundance formats. The authors review the zonal-mean ozone variations measured over the two and a half years since launch in September 1991. Well-known features such as the annual and semiannual variations are ubiquitous. In the equatorial regions, longer-term changes are believed to be related to the quasi-biennial oscillation (QBO), with a strong semiannual signal above 20 hPa. Ozone values near 50 hPa exhibit an equatorial low from October 1991 to June 1992, after which the low ozone pattern splits into two subtropical lows (possibly in connection with residual circulation changes tied to the QBO) and returns to an equatorial low in September 1993. The ozone hole development at high southern latitudes is apparent in MLS column data integrated down to 100 hPa, with a pattern generally consistent with Nimbus-7 Total Ozone Mapping Spectrometer (TOMS) measurements of total column; the MLS data reinforce current knowledge of this lower-stratospheric phenomenon by providing a height-dependent view of the variations. The region of 30-degrees-S to 30-degrees-N (an area equal to half the global area) shows very little change in the ozone column from year to year and within each year.The most striking ozone changes have occurred at northern midlatitudes, with the October 1992 to July 1993 column values significantly lower than during the prior year. The zonal-mean changes manifest themselves as a slower rate of increase during the 1992/93 winter, and there is some evidence for a lower fall minimum. A recovery occurs during late summer of 1993; early 1994 values are significantly larger than during the two previous winters. These results are in general agreement with variations measured by the Nimbus-7 TOMS and Meteor-3 TOMS instruments at midlatitudes. However, the southern midlatitudes exhibit less of a column ozone decrease (relative to the north) in the MLS data (down to 100 hPa) than in the TOMS column results. The timing and latitudinal extent of the northern midlatitude decreases appear to rule out observed ClO enhancements in the Arctic vortex, with related chemical processing and ozone dilution effects, as a unique cause. Local depletion from ClO-related chemical mechanisms alone is also not sufficient, based on MLS ClO data. The puzzling asymmetric nature of the changes probably requires a dynamical component as an explanation. A combination of effects (including chemical destruction via heterogeneous processes and QBO phasing) apparently needs to be invoked. This dataset will place constraints on future modeling studies, which are required to better understand the source of the observed changes.Finally, residual ozone values extracted from TOMS-minus-MLS column data are briefly presented as a preliminary view into the potential usefulness of such studies, with information on tropospheric ozone as an ultimate goal.

Observations of ozone from the Microwave Limb Sounder (MLS) aboard the Upper Atmosphere Research Satellite (UARS).

The Microwave Limb Sounder observed waves in stratospheric temperature and O3 during the 1992 southern winter. Wave 1 intensifies three times from mid August through mid September, when a 9 day eastward traveling wave becomes in phase with the stationary wave 1. During the periods of wave intensification, minor sudden warmings and increased zonal mean O3 are observed. The waves have a westward phase tilt which results in an intensified baroclinic zone when the waves are in phase. Waves in T and O3 are positively correlated near 5–10 hPa, implying transport by planetary waves; this is supported by larger O3 wave amplitudes than expected from photochemistry alone.

Fourier analysis has been applied to data obtained from limb viewing instruments on the Upper Atmosphere Research Satellite. A coordinate system rotation facilitates the efficient computation of Fourier transforms in the temporal and longitudinal domains. Fields such as ozone (O3), chlorine monoxide (ClO), temperature, and water vapor have been transformed by this process. The transforms have been inverted to provide maps of these quantities at selected times, providing a method of accurate time interpolation. Maps obtained by this process show evidence of both horizontal and vertical transport of important trace species such as O3 and ClO. An examination of the polar regions indicates that large‐scale planetary variations are likely to play a significant role in transporting midstratospheric O3 into the polar regions. There is also evidence that downward transport occurs, providing a means of moving O3 into the polar vortex at lower altitudes. The transforms themselves show the structure and propagation characteristics of wave variations.

UARS MLS measurements of ClO in the 1992 southern hemisphere winter are described. Lower stratospheric ClO abundances greater than 1 ppbv were observed in the vortex beginning 1 June. The enhanced ClO reached largest areal extent in mid-August, then retreated poleward. ClO abundances at 22 hPa decreased in early September while those at 46 hPa remained high. O3 decrease within the vortex was observed by mid-August, and was coincident with the enhanced ClO.

Concentrations of atmospheric ozone and of CIO (the predominant form of reactive chlorine responsible for stratospheric ozone depletion) are reported for both the Arctic and Antarctic winters of the past 18 months. Chlorine in the lower stratosphere was almost completely converted to chemically reactive forms in both the northern and southern polar winter vortices. This occurred in the south long before the development of the Antarctic ozone hole, suggesting that ozone loss can be masked by influx of ozone-rich air.

The evolution of ozone observed by UARS MLS in the Northern Hemisphere polar vortex is shown as a function of time throughout the stratosphere, for the 1991-1992 and 1992-1993 winters.

The evolution of ozone (O3) observed by the Microwave Limb Sounder on board the Upper Atmosphere Research Satellite is described for 14 Aug through 20 Sep 1992, in relation to the polar vortex. The development of an ozone hole is observed in column O3, and a corresponding decrease is seen in O3 mixing ratio in the polar lower stratosphere, consistent with chemical destruction. The observations also suggest that poleward transport associated with episodes of strong planetary wave activity is important in increasing O3 in the mid‐stratosphere.